The Pleiades doesn't look right.
The Pleiades doesn't look right.
The Pleiades star cluster in Celestia doesn't look right. It looks right from the Earth viewpoint, but then when I circle around the Pleiades the stars take on a strange pattern; the length of the cluster stretches way out of proportion to its width. I'm sure the real Pleiades isn't arranged that way. Something is miscalculated with those star distances in Celestia.
-
Evil Dr Ganymede
- Posts: 1386
- Joined: 06.06.2003
- With us: 21 years 8 months
Well, the distance between Pleione and Celaeno (the furthest and closest of the named Pleiades) is about 53 ly according to Celestia.
There is a 3D Hyades that shows that the stars in that cluster are spread out over quite a large distance (it's basically a cylinder 100 pc long and about 40 pc wide).
The Hyades are an older, more wide-open cluster than the Pleiades though. Considering that the bright stars in the Pleiades seem to be contained in a sphere about 20 pc in diameter, I suspect that the values in Celestia are actually correct.
So while it does seem contrary to what you might expect, I don't think the distances are wrong. (though I don't know whether this article about the distance to the Pleiades may be relevant here).
While I'm here... is there a complete list of data for all the stars (not just the obvious "seven sisters" that belong to the Pleiades cluster)?
There is a 3D Hyades that shows that the stars in that cluster are spread out over quite a large distance (it's basically a cylinder 100 pc long and about 40 pc wide).
The Hyades are an older, more wide-open cluster than the Pleiades though. Considering that the bright stars in the Pleiades seem to be contained in a sphere about 20 pc in diameter, I suspect that the values in Celestia are actually correct.
So while it does seem contrary to what you might expect, I don't think the distances are wrong. (though I don't know whether this article about the distance to the Pleiades may be relevant here).
While I'm here... is there a complete list of data for all the stars (not just the obvious "seven sisters" that belong to the Pleiades cluster)?
Remember that Celestia uses the distances measured by the Hipparcos astrographic satellite. Unfortunately, some of its parallax measurements have very large error bars. The phrase "large error bars" means that they aren't very accurate.
The errors in the measurements of the distances of a group of stars that really are close to one another will make the group seem to be stretched out, which is what you're seeing.
To get an idea of just how gross some of the errors are, I created an Addon that draws lines of +/- 1 sigma through the star positions. "1 sigma" indicates that there's about a 68 percent probability that the star's actual position is somewhere on that line, and a 32 percent probability that its real position is even further away from where it's drawn.
See http://www.lns.cornell.edu/~seb/celestia/catalogs.html#3.5.8
FWIW, there is a controversy about the distances that Hipparcos measured for the Pleiades and Beehive clusters, which has not yet been resolved.
The WebDA Open Cluster database at http://obswww.unige.ch/webda/ lists members of many clusters. Its information about the Pleiades is at http://obswww.unige.ch/webda/cgi-bin/ocl_page.cgi?cluster=pleiades
Select the "available data" link to learn more than you want
In particular, it lists more than 600 stars with > 90% probability of being cluster members.
The errors in the measurements of the distances of a group of stars that really are close to one another will make the group seem to be stretched out, which is what you're seeing.
To get an idea of just how gross some of the errors are, I created an Addon that draws lines of +/- 1 sigma through the star positions. "1 sigma" indicates that there's about a 68 percent probability that the star's actual position is somewhere on that line, and a 32 percent probability that its real position is even further away from where it's drawn.
See http://www.lns.cornell.edu/~seb/celestia/catalogs.html#3.5.8
FWIW, there is a controversy about the distances that Hipparcos measured for the Pleiades and Beehive clusters, which has not yet been resolved.
The WebDA Open Cluster database at http://obswww.unige.ch/webda/ lists members of many clusters. Its information about the Pleiades is at http://obswww.unige.ch/webda/cgi-bin/ocl_page.cgi?cluster=pleiades
Select the "available data" link to learn more than you want
In particular, it lists more than 600 stars with > 90% probability of being cluster members.Selden
-
Evil Dr Ganymede
- Posts: 1386
- Joined: 06.06.2003
- With us: 21 years 8 months
There's something on the WebDA page that says that the "tidal radius" of the Pleiades is 16 pc, which is close to the distance between the two stars I found earlier:
http://obswww.unige.ch/webda/abstracts/ ... iades.html
Dunno exactly what "Tidal Radius" means here, but presumably that's something to do with how well the stars can gravitationally interact with eachother?
http://obswww.unige.ch/webda/abstracts/ ... iades.html
Dunno exactly what "Tidal Radius" means here, but presumably that's something to do with how well the stars can gravitationally interact with eachother?
If it helps, the full text of the paper is available from http://arxiv.org/ by searching the astro-ph category for papers by Raboud.
Specifically, see http://arxiv.org/PS_cache/astro-ph/pdf/9708/9708144.pdf
.
Specifically, see http://arxiv.org/PS_cache/astro-ph/pdf/9708/9708144.pdf
.
Selden
Just now I looked at the error bars on the Pleiades members drawn by Celestia. Their error bars are as large as or larger than the size of the apparent cluster. In other words, although the distance estimates for the cluster as a whole may be about right, the relative distances of the individual members cannot be trusted.
So it does seem reasonable to readjust the distances shown for the cluster members, so long as this adjustment is clearly documented. In principle, other methods can be used to determine their relative distances fairly accurately. I'm sure someone has done this already.
So it does seem reasonable to readjust the distances shown for the cluster members, so long as this adjustment is clearly documented. In principle, other methods can be used to determine their relative distances fairly accurately. I'm sure someone has done this already.
Selden